KR101523909B1 - Method and apparatus to configure thermal design power in a microprocessor - Google Patents

Abstract

A technique for changing a thermal design power (TDP) value is disclosed. In one embodiment, one or more environments or user-driven changes may cause the TDP value of the processor to change. Moreover, in some embodiments, changing the TDP may change the turbo mode target frequency.

Description

[0001] METHOD AND APPARATUS TO CONFIGURE THERMAL DESIGN POWER IN A MICROPROCESSOR [0002]

Related application

This application is a continuation-in-part of U.S. Patent Application No. 12 / 975,100, filed December 21, 2010, now pending.

Field of invention

Embodiments of the present invention generally relate to the field of information processing, and more specifically to the field of power management of computing systems and microprocessors.

Control of power consumption of microprocessors is increasing in importance. Conventional techniques for controlling processor power consumption do not adequately allow a resilient configuration of the thermal design power (TDP) specification for the processor.

1 illustrates a technique for configuring thermal design power (TDP) in accordance with one embodiment;
2 illustrates an embodiment of at least one technique for constructing a TDP according to an embodiment.
3 illustrates aspects of an initialization technique corresponding to a configurable TDP in accordance with one embodiment;
4 illustrates an embodiment of at least one technique for constructing a TDP according to an embodiment.
5 illustrates an embodiment of at least one technique for constructing a TDP according to an embodiment.
6 illustrates an embodiment of at least one technique for constructing a TDP according to one embodiment;
Figure 7 is a block diagram of a microprocessor in which at least one embodiment of the present invention may be used.
8 is a block diagram of a shared bus computer system in which at least one embodiment of the present invention may be used.
Figure 9 is a block diagram of a point-to-point interconnection computer system in which at least one embodiment of the present invention may be used.

Embodiments of the invention are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings in which like reference numerals represent like elements.

An embodiment of the present invention relates to a configured thermal design power (TDP) for a processor. While there are many embodiments of the many aspects of the present invention, at least one of the aspects is exemplified herein as an example for teaching the embodiments of the present invention and should not be construed as being entirely comprehensive and mutually exclusive.

A processor may be rated or specified to include both performance and power related characteristics. An individual product or family of products may have an associated specification that includes specific base and turbo frequency capabilities as well as other performance related characteristics. The range of power consumption of the processor may be specified for the product family. For example, a standard voltage (SV) mobile processor may have a thermal design power (TDP) rating of 35 watts. This rating may be an indication to the OEM that the processor purchased by the original equipment manufacturer (OEM) when executed in the specified TDP workload may dissipate power less than or equal to the specified TDP value for the product, At worst, when operating at temperatures, the worst case real-world workload scenarios can be expressed. While the specified performance characteristics may change across the product family, the TDP may be specified as the same value across a number of product families. This allows OEMs to design a single platform that can dissipate a specified TDP while providing a range of performance at different price points. On the other hand, the turbo capability is a potential performance advantage in that the TDP workload allows the TDP power to be dissipated at the fundamental frequency.

In some embodiments, there are multiple types of turbo modes. Above is a version of Turbo mode that belongs to either workload or package power sharing, in which case the workload with a lower power inherent to the TDP apps that are otherwise turbomeless can benefit from the same frequency. These workloads can be benefited by providing more power than the fundamental frequency, thereby raising the power back to the package TDP power. The other version of the turbo mode is a dynamic turbo mode where power is allowed to exceed the TDP power for a limited period of time so that on average the power is still TDP power over time and this is for example an idle workload or only TDP power Regardless of whether it is a workload that does not draw the same power to the threshold, priority over power less than TDP allows a slight deviation in excess of TDP. TDP power also affects power delivery design needs.

Platform thermal capability is a design choice for OEMs as this affects size, weight, audible noise, and material bill of materials (BOM) costs. Thus, the specified TDP for the processor may have a significant impact on the platform thermal design. Some processors are marketed with only a small number of TDPs. For example, the SV of a mobile processor is 35W, the SV of a low-voltage processor is 25W, and the SV of an ultra-low-voltage processor is 18W.

In a mobile platform, cooling capability and audible noise tolerance can vary depending on the usage environment. For example, when docked and executed in alternating current (AC) power, the platform may have more cooling capability than when it is run undocked to battery power.

Running a high power processor in an environment with less cooling capability than capable of cooling a specified TDP can reduce performance by having thermal control actually at an unknown capability level that varies across the product. Additionally, the turbo capability may be disabled as part of the thermal control.

The configurable TDP may, according to one embodiment, allow the OEM to configure the processor's TDP with one of a number of values. This configuration can be performed statistically or dynamically at " on-the-fly "at initialization. This is effectively accomplished by changing the fundamental frequency of the processor to one of a number of supported fundamental frequencies. The meaning of this change may be that a particular performance is guaranteed by the fundamental frequency and the TDP is specified for each of the supported fundamental frequencies. Additionally, when the fundamental frequency / TDP is changed, the point at which the turbo is coupled can also be changed accordingly. This provides the OEM with the ability to ensure that maximum power dissipation is known while still delivering turbo boost performance when the provided workload allows it.

Figure 1 shows a "TDP up" 105, a "nominal TDP" 110, and a "TDP down " 110, which may correspond to power states such as P-states P0 130, P5 125 and P9 120, "≪ / RTI > FIG. 11 illustrates a technique for providing a configurable TDP to a processor, according to one embodiment, by providing three TDP levels, Because the TDP value is changed dynamically, in one embodiment, the amount of turbo capability is similarly modified to allow even more rise to turbo boost technology while still providing specific performance to the end user.

In one embodiment, the configurable TDP technique includes an enabled and configured set of frequency and TDP values provided for the processor. In one embodiment, the validated values are fused in hardware, programmed, or otherwise configured to enable platform firmware or software to detect and use capabilities.

In one embodiment, the configurable TDP provides a mechanism for designing a processor as a new segment of the platform. For example, a processor that supports a configurable TDP may have a premium over other non-TDP processors. The OEM can then choose to purchase the processor and configure it for their needs, or to provide it to a platform that supports on-the-fly reconfiguration of performance and power. One such example is the "Mobile Extreme Edition" platform on a non-docked battery. The configurable TDP likewise has the potential to reduce the number of product families provided.

In one embodiment, the configurable TDP architecture does not make assumptions about standards, interdependencies with other technologies, and so on. The following Table 1 illustrates various aspects and portions of a platform that may be affected by the configurable TDP according to one embodiment.

In one embodiment, a change may be made to the processor's model specific register (MSR), and the new MSR may be used to support a configurable TDP for the processor. These registers can provide the ability to change the point at which the turbo is coupled and set the Run Time Average Power Limit (RAPL) power limit for the new fundamental frequency. In one embodiment, the list of registers that can be accessed, modified or added using the configurable TDP includes:

PLATFORM_INFO: This register can be used to detect configurable TDP capabilities.

CONFIG_TDP_LIMIT_1; CONFIG_TDP_LIMIT_2: This register can be used to detect configurable TDP ratio and corresponding TDP power and power range.

CONFIG_TDP_CONTROL: This register can be used by software to select different TDP point and read current selection.

PSTATE_NOTIFY Hook: This register can be used by software to enable turbo from the new P1 point. Separating this register from CONFIG_TDP_CONTROL may allow a usage model that allows the OS to select a specific sealing for an acceptable turbo range.

In other embodiments, other registers or storage devices (e.g., memory, cache, etc.) may be used to provide the configurable TDP. Moreover, in some embodiments, the functions provided in the registers may be incorporated into fewer registers or storage devices.

In some embodiments, there may not be a unique platform physical requirement for the configurable TDP. However, in some embodiments, the specifications for power transfer and cooling may be developed to address the requirement for each of the TDP points. The specification may reflect the ability to select the TDP level for the design and not accept or accept other points in some embodiments.

In some embodiments, a new interface or technology may not be specifically required to support a configurable TDP. However, in some embodiments, the design characteristics that can be addressed and enabled in the specification include the thermal design current (ITDC) and the maximum possible current that can be supported (e.g., "Iccmax"). In some embodiments, the parameters may be defined for each TDP point.

The TDP may imply an indefinitely sustained cooling level to be presented to support the corresponding TDP power level. However, in one embodiment, no particular technique is required to express a change in cooling capability, whether it is exotic design, docked cooling, change in fan speed, change in ambient environment, and the like. However, the cooling design parameters can be set for each TDP level in enabling documentation.

In one embodiment, a configurable TDP for other logic such as graphics, memory control, or peripheral control may be used. For example, if a configurable TDP is used for graphics, the graphics driver may need to be notified about the new TDP level and the corresponding RP1 frequency. This may be done in at least two ways in one embodiment.

(1) via the interrupt from the processor to the graphics device driver when the TDP level and the corresponding RP1 frequency are changed. Which in one embodiment may require an interrupt configuration and status register in addition to the registers already required to support the configurable TDP.

(2) via the software stack notifying the graphics driver when the TDP level and corresponding RP1 frequency are changed. This may require updating to the software-to-graphics driver communication interface already in place as part of the software stack.

In one embodiment, the TDP configuration allows the OS to use a specific P-state (e.g., an ACPI notification), expose all possible P-states to the OS at initialization, enable turbo capability at various operating points You can ask the platform to stop doing that. The ACPI P-State Table (PSS) may be suitably populated in some embodiments. In one embodiment, there may not be an ecosystem requirement to support the configurable TDP.

In one embodiment, the configurable TDP is statically configured by the BIOS during initialization, e.g., dynamically during the run time, to a value other than the default fused by the BIOS or software driver. In one embodiment, configuring the TDP records a new turbo ratio limit in the MSR and sets the corresponding power limit to the RAPL power limit (MSR) according to the value specified for the partial / fundamental frequency and by setting the point at which the turbo is coupled ≪ / RTI > Additionally, in some embodiments, the operating system may be notified to limit its use of the P-state according to the new fundamental frequency. This can be accomplished in one embodiment by allowing the OS to evaluate the ACPI_PPC object (performance presentation capability) under each logical processor.

Figure 2 illustrates logic for configuring the TDP in accordance with one embodiment. The logic shown in FIG. 2 may be included in the processor hardware or some other hardware. Alternatively, the logic of FIG. 2 may be incorporated into a tangible machine readable medium in which instructions that, when executed, enable the functionality of the logic shown in FIG. 2 to be performed are stored therein. In FIG. 2, the OSPM power configuration applet 215 may be optional, and its use of DPPE functions as a trigger 245 to invoke a TDP configuration change. At the trigger 245, for example, upon changing the power source or power plan, the applet communicates the change to the DPTF driver 220.

The DPTF driver 220 receives the TDP configuration changes from the OSPM power configuration applet and as a result performs two functions. The first is to evaluate the ACPI object within its device category that notifies it to re-evaluate the PPC object under each logical processor by causing the ACPI notification to be issued by the BIOS 210 on the logical processor 225 to the OS. The return value from the object is derived from the value passed by the DPTF driver 220 and limits the use of the particular P-state's operating system below the new fundamental frequency. After this is accomplished, the DPTF driver 220 writes a new TDP configuration 270 to the processor 225 (MSR recording) to determine the new turbo ratio for the processor (the point at which the turbo is called) Set the RAPL power limit.

In one embodiment, the processor includes the MSR described above. Writing the MSR can deliver RAPL power limit values as well as information to the power control unit (PCU) in the P-state where the turbo is called (turbo ratio).

In one embodiment, the BIOS 210 includes ACPI firmware and natively executable code. The BIOS 210 may be responsible for detecting the configurable TDP 255 feature availability in one embodiment and for properly setting up the ACPI firmware structure (_PSS). BIOS 210 may statically configure the TDP to be less than the maximum value for the product or product family during initialization. Alternatively, the BIOS 210 alone can dynamically configure the TDP configuration through a combination of SMM execution and ACPI notification in one embodiment. The DPTF may also be used to dynamically set the TDP configuration, but in some cases the BIOS 210 may include ACPI firmware that is evaluated to signal the OS to reevaluate the _PPC object under each logical processor. In one embodiment, the _PPC object evaluation determines which P-state is currently available for use by the OS corresponding to the TDP configuration (including the P-state in which the turbo is invoked).

In one embodiment, the OS receives an ACPI notification 230 that causes the _PPC object to be reevaluated under each logical processor. The value returned from the evaluation of the _PPC object restricts the operating system P-state control (205) software from using a particular P-state according to the TDP configuration. When the TDP configuration is changed, the highest performance P-state allowed by the _PPC object is configured to be the P-state that invokes the turbo operation.

To initialize the configurable TDP 255, the platform BIOS 210 may first detect feature availability according to one embodiment. It can then build the OSPM_PSS table using the configurable TDP (255) information it collects from the processor. Figure 3 illustrates an initialization technique in accordance with one embodiment. For example, the BIOS 310 may first detect feature availability within the process 330. The BIOS 310 may then construct the OSPM_PPS table in the process 325 using the configured TDP information, such as the TDP level and ratio, which it collects from the processor / PCU 315 in process 335.

The BIOS 310 may program the maximum TDP ratio or the desired TDP ratio as the current TDP ratio within the processor / PCU 315 as illustrated in process 340. [ BIOS 310 may also set the P-state corresponding to the desired TDP ratio or _PPC to zero "0 " to indicate the allowed P-state in process 345 and report the _PPS table to OSPM 320 . OSPM 320 may change the P-state for processor / PCU 315 to the new maximum P-state (in accordance with the workload) in process 355. [ If the target rate is greater than the current P1 rate, the processor / PCU 315 may enable the turbo in process 350. The initialization techniques of FIG. 3 and other processes or mechanisms disclosed herein may be implemented by processing logic that may include dedicated hardware or software or firmware opcode executable by a general purpose machine, by a special purpose machine, or by a combination of the two. .

In one embodiment, there are three possible mechanisms by which the TDP may change during runtime. In other embodiments, other techniques or mechanisms may be used to change the TDP during execution time. In one embodiment, the platform may provide an option for the end user 405 to select a particular mode of operation for the system, which may be provided as a hotkey input 425. In this example, the hotkey operation by the user triggers a change in the TDP value during execution time. 4 illustrates a flow of a user initiated TDP change in accordance with one embodiment. After the selection of a new TDP mode of operation for the system as a hotkey input in process 425, the BIOS 410 programs the new P1 ratio and sends a new TDP point to the new TDP point in the processor / PCU 415 as illustrated in process 430 RAPL power limit can be programmed. The BIOS 410 may also set the _PPC from the new maximum available P-state (_PSS to the new turbo P-state) in process 440 and report the _PPC table to the OSPM 420. The OSPM 420 may change the P-state for the processor / PCU 415 to the new maximum P-state (in accordance with the workload) in process 445. [ If the target rate is greater than the current P1 rate, the processor / PCU 415 may enable turbo in process 435. [

The usage model, in accordance with one embodiment, uses platform software to intercept user input and translate it into a BIOS 510 call to invoke a TDP change. Figure 5 shows this usage model according to one embodiment. In process 530, for example, event 505 may include user selection of a new TDP mode via power-fan settings or a software GUI or dock, and the change triggered by event 505 may include a DPTF Driver < / RTI > The DPTF driver 520 issues a new P 1 notification that notifies the ACPI notification to be issued by the BIOS 510 in process 540 to set the _PPC object to the new maximum available P-state (_PSS to new turbo P-state) Optionally, call the ACPI method. The OSPM 515 may then change the P-state for the processor / PCU 515 to the new maximum P-state (in accordance with the workload) in process 555. [ The return value from the ACPI object is derived from the value passed by the DPTF driver 520.

In one embodiment, the processor / PCU 525 includes an MSR as described above. Writing the MSR can deliver the RAPL power limit as well as information to the processor / PCU in the (turbo) P-state where the turbo is invoked. Thus, at process 545, the DPTF driver 520 programs the processor / PCU 525 to a new P1 ratio (via MMIO / MSR recording) to generate a new turbo ratio for the processor / PCU 525 , And programs the corresponding RAPL power limit for the new TDP base frequency point. If the target rate is greater than the current P1 rate, the processor / PCU 525 may enable turbo in process 550.

In some embodiments, the platform does not provide user control to modify the TDP, but the platform may choose to base decisions on system events such as AC-to-DC switches, or docked vs. non-docked events, and the like. This usage model is shown in the sequence shown in FIG. 6 according to one embodiment. In process 625, for example, platform EC 605 notifies BIOS 610 of a new TDP request in accordance with the system event, as described above. The BIOS 610 may program the new P1 ratio and program the RAPL power limit to the new TDP point in the processor / PCU 615 as illustrated in process 630. [ The BIOS 610 may set the _PPC to the new maximum available P-state (the new turbo P-state in the _PSS) and report the _PSS table to the OSPM 620 in process 635. The OSPM 620 may then change the P-state for the processor / PCU 615 to the new maximum P-state in process 645 (depending on the workload). If the target rate is greater than the current P1 rate, the processor / PCU 615 may enable turbo in process 640. [

In one embodiment, the TDP configuration may be changed dynamically as described above.

In one embodiment, the configurable TDP may interoperate with platform firmware and thermal control capabilities to manipulate ACPI objects to ensure that conflicts do not occur. In one embodiment, the run time average power limit (RAPL) allows the platform to limit the power consumption of the processor. Because the platform can use the TDP details based on the RAPL limits, the fact that the TDP changes dynamically may cause the RAPL limit to be invalidated. For example, consider the case where the current TDP is 15 W and the RAPL limit is set to 14 W by the platform. When the current TDP is changed to 23 W, the RAPL limit of 14 W is too restrictive and the processor will not be able to maintain the RAPL limit. To solve this problem, the RAPL limit may be updated as part of the configurable TDP change during run time to match the new TDP level according to one embodiment.

In one embodiment, the configurable TDP maps to two platform features (grouping of interface characterizations). These are the TDP configurable interfaces (configuration) and triggers.

Some additional features include, in accordance with one embodiment, new processor MSRs and graphics driver changes or new uses thereof.

Figure 7 illustrates a microprocessor in which at least one embodiment of the present invention may be used. In particular, FIG. 7 illustrates a microprocessor 700 having one or more processor cores 705, 710, each associated with a respective local cache 707, 713. 7 also shows a shared cache memory 715 that can store at least some versions of the information stored in each of the local caches 707,713. In some embodiments, microprocessor 700 also includes other logic not shown in FIG. 7, such as an integrated memory controller, an integrated graphics controller, as well as other logic for performing other functions in a computer system such as I / O control. . In one embodiment, each processor core in a respective microprocessor or multi-core processor in a multi-processor system includes logic 719 to enable a resilient configuration of the TDP specification technique, according to at least one embodiment. Or otherwise associated. The logic may include circuitry, software (embodied in a tangible medium), or both, to enable resource allocation among a plurality of cores or processors that is more efficient than some conventional implementations.

Figure 8 illustrates a front side bus (FSB) computer system, for example one embodiment of the present invention may be used. Any one of the local level 1 (L1) cache memories (801, 805, 810, or 815) associated within or in one of the processor cores 823, 827, 833, 837, 843, 847, 853, (820, 825, 830, 835, 840, 845, 850, 855). Further, any processor 801, 805, 810 or 815 may be implemented from any one of the shared Level 2 (L2) caches 803, 807, 813, 817 or from the system memory 860 via the chipset 865 To access the information. One or more of the processors of FIG. 8 may include or otherwise be associated with logic 819 to enable resilient configuration of the TDP specification technique in accordance with at least one embodiment.

In addition to the FSB computer system shown in FIG. 8, other system configurations, including point-to-point (P2P) interconnect systems and ring interconnect systems, can be used with various embodiments of the present invention. The P2P system of FIG. 9 may include, for example, multiple processors, of which only two processors 970 and 980 are shown by way of example. Processors 970 and 980 may each include a local memory controller hub (MCH) 972 and 982 for connecting with memories 92 and 94, respectively. Processors 970 and 980 may exchange data via point-to-point (PtP) interface 950 using PtP interface circuits 978 and 988. [ Processors 970 and 980 may exchange data with chipset 990 via respective PtP interfaces 952 and 954 using point-to-point interface circuits 976, 994, 986 and 998, respectively. The chipset 990 may also exchange data with the high performance graphics circuit 938 via the high performance graphics interface 939. Embodiments of the present invention may be located within each of the PtP bus agents of FIG. 9, or within any processor having any number of processing cores. In one embodiment, any processor core may include or be associated with a local cache memory (not shown). Moreover, a shared cache (not shown) may be included in any one processor external to both processors and still be connected to the processor via a P2P interconnect so that when the processor is placed in a low power mode, Local cache information can be stored in the shared cache. One or more of the processors or cores of FIG. 9 may include or otherwise be associated with logic 919 to enable a resilient configuration of the TDP specification technique, according to at least one embodiment.

One or more aspects of at least one embodiment may be implemented as a machine that expresses or combines various functional and / or logical data and / or logic within a processor and, when read by the machine, causes the machine to produce logic to perform the techniques described herein And can be implemented by expressive data stored on a readable medium. This representation, known as "IP core ", is stored on a tangible machine readable medium (" tape ") and can be supplied to various customers or manufacturing facilities to load into a manufacturing machine that actually produces the logic or processor.

Embodiments of the present invention may be embodied or applied in any hardware device or portion thereof, including a central processing unit, a graphics processing unit or other processing logic or core within a processor or in a computer system. Embodiments may also be embodied in a tangible machine readable medium having stored thereon a set of instructions that, when executed by a machine, cause the machine to perform the operations described herein.

Thus, a method and apparatus for directing micro-architecture memory area access has been described. It is to be understood that the above description is intended to be illustrative, not limiting. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. Accordingly, the scope of the invention may be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims (24)

A processor that supports configuring thermal design power (TDP)
One or more model specific registers (MSRs) supporting a configurable TDP specific to the processor,
Logic for changing a configurable TDP value in response to a user control trigger provided by the platform, the change comprising updating a runtime average power limit (RAPL) value as part of a configurable TDP change - containing
Processor.
The method according to claim 1,
The logic uses the platform software to intercept the user control trigger and convert it to basic input / output software (BIOS) calls to cause the configurable TDP change
Processor.
3. The method of claim 2,
The logic is for changing the configurable TDP value in response to a power change from alternating current (AC) to direct current (DC) or from a DC to AC power change
Processor.
3. The method of claim 2,
Wherein the logic is for changing the configurable TDP value in response to a docking event, the change comprising a change in a turbo ratio value
Processor.
The method according to claim 1,
Wherein the configurable TDP value is changed from an initially configured TDP value in response to the user control trigger
Processor.
The method according to claim 1,
Wherein the configurable TDP value is modified in response to storing information in at least one model specific register (MSR)
Processor.
The method according to claim 1,
Changing the configurable TDP value causes a corresponding change in the turbo mode target frequency
Processor.
The method according to claim 1,
Wherein said configurable TDP value is changed in response to a change in ambient temperature
Processor.
A system that supports configuring thermal design power (TDP) in a processor,
A processor, comprising: a plurality of logical processors; and one or more model specific registers (MSRs) supporting a configurable TDP specific to the processor;
Logic for changing a configurable TDP value in response to a user control trigger, said modification comprising updating a run time average power limit (RAPL) value as part of a configurable TDP change;
And a memory for storing instructions to be executed by the processor
system.
10. The method of claim 9,
The logic uses the platform software to intercept the user control trigger and convert it to basic input / output software (BIOS) calls to cause a configurable TDP change
system.
11. The method of claim 10,
The logic is for changing the configurable TDP value in response to a power change from alternating current (AC) to direct current (DC) or from a DC to AC power change
system.
11. The method of claim 10,
Wherein the logic is for changing the configurable TDP value in response to a docking event, the change comprising a change in the turbidity value
system.
10. The method of claim 9,
Wherein the configurable TDP value is changed from an initially configured TDP value in response to a user control trigger
system.
10. The method of claim 9,
The configurable TDP value is changed in response to storing information in at least one model specific register (MSR)
system.
10. The method of claim 9,
Changing the configurable TDP value causes a corresponding change in the turbo mode target frequency
system.
10. The method of claim 9,
Wherein the configurable TDP value is altered in response to a change in ambient temperature, heat, acoustic or system environmental conditions
system.
A method of configuring thermal design power (TDP) in a processor,
Changing a configurable processor TDP value in response to a user control through one or more model specific registers (MSRs) for the processor, the change comprising a runtime average power limit RAPL) values,
Comprising: causing the operating system to evaluate an Advanced Configuration and Power Interface (ACPI) Performance Present Capabilities (PPC) object under each of the plurality of logical processors as part of the configurable TDP change
Way.
18. The method of claim 17,
Further comprising using platform software to intercept the user control and convert it to basic input / output software (BIOS) code to cause the configurable TDP change
Way.
19. The method of claim 18,
Further comprising changing the configurable TDP value in response to a power change from AC to DC or DC to AC power change
Way.
19. The method of claim 18,
Further comprising changing the configurable TDP value in response to a docking event, the change comprising a change in the turbidity value
Way.
18. The method of claim 17,
Wherein the configurable TDP value is changed from an initially configured TDP value in response to the user control
Way.
18. The method of claim 17,
Wherein the configurable TDP value is modified in response to storing information in at least one model specific register (MSR)
Way.
18. The method of claim 17,
Changing the configurable TDP value causes a corresponding change in the turbo mode target frequency
Way.
18. The method of claim 17,
Wherein said configurable TDP value is changed in response to a change in ambient temperature
Way.

Images